Fault Classification and Location in Microgrid Using Artificial Neural Networks

Abstract

—A microgrid is a compact, localized power sys tem that independently generates, distributes, and regulates electricity, either standalone or in sync with the main grid. These microgrids are designed to ensure a dependable power supply to specific areas. Intelligent microgrids have been made possible through the use of advanced sensors and the most recent grid communication standards. Nonetheless, when utilized in microgrids, conventional protection methods do not yield dependable results. This article presents a technique that employs measurements of three-phase voltage, current, and angle during a fault as input data for a module that classifies and locates faults. This module, constructed using an artificial neural network (ANN) technique, is part of the central protection system. The effectiveness of the suggested approach is evaluated by taking into account actual grid situations with different fault locations and types. A 7-bus meshed AC Microgrid Test System, which includes two Distributed Generators (DGs) and two grid sources, is simu lated in the Simulink platform. MATLAB-2021b’s data analytic capabilities have been utilized for the development of ANN-based fault classification and location modules for microgrids.